Cloudflare Says Bots Could Generate 1,000x More Web Traffic Than Humans
Cloudflare says AI agents, bots, and other automated systems could generate up to 1,000 times more internet traffic than humans within five years as machine activity accelerates. Photo: HaeB / WiKi
AI & Machine Learning

Cloudflare Says Bots Could Generate 1,000x More Web Traffic Than Humans

Cloudflare says AI agents, bots, and other automated systems could generate up to 1,000 times more internet traffic than humans within five years as machine activity accelerates.

By Daniel Mercer • 4 mins read Edited by AIstify Team Published: Updated:

Key Notes

  • Cloudflare CFO Thomas Seifert predicts non-human internet traffic could become as much as 1,000 times larger than human traffic within five years.
  • Machine-generated traffic already surpassed human traffic in May 2026, earlier than Cloudflare’s previous forecast of 2027.
  • Cloudflare expects the machine-traffic surge to increase infrastructure demands and expand the cybersecurity threat surface.

Cloudflare says humans could become almost statistically invisible in internet traffic within five years as AI agents, crawlers, bots, and other automated systems generate vastly more requests than people.

Cloudflare CFO Thomas Seifert said non-human traffic could become as much as 1,000 times larger than human traffic if current growth trends continue. The forecast was made during the company’s second-quarter earnings call and was reported by The Register.

“Humans will be a rounding error on the internet,” Seifert said, while stressing that the shift would not come from people using the web less. Instead, Cloudflare expects automated activity to grow so quickly that human-generated traffic becomes a tiny fraction of the total.

The milestone is already arriving faster than the company anticipated. Cloudflare had previously expected machine traffic to overtake human activity in 2027. Its measurements showed the crossover happened in May 2026.

Cloudflare later said automated systems were responsible for roughly 57% of all web requests, while a separate company analysis found that more than half of internet traffic was already non-human. The trend is closely tied to the rapid expansion of AI agents that continuously search, retrieve, analyze, and act on online information.

One Human Request Can Trigger Hundreds of Machine Actions

The economics of agentic computing are fundamentally different from ordinary web browsing. A person may open a handful of pages to research a purchase or answer a question. An AI agent can perform dozens or hundreds of requests in the background while comparing products, collecting information, calling APIs, checking availability, and completing a task.

That amplification means human internet use can continue growing even as its share of total traffic falls sharply. Cloudflare’s July analysis said more than 50% of traffic was already non-human, with 52% of identified crawler requests being used for AI training as of June. Mixed-purpose crawlers combining search, training, and agent functions accounted for another 36% of crawler activity.

The change is especially important for the telecoms and networks industry because automated traffic still consumes bandwidth, compute, storage, caching capacity, and security resources regardless of whether a human directly sees the resulting page.

Cloudflare says websites will increasingly need to treat machines as a second audience. During its Agents Week announcements, the company said fewer than half of HTML page requests now come directly from humans and introduced tools that show site owners whether AI systems can successfully read, understand, and recommend their content.

This shift also changes how websites make money. Traditional search crawlers generally indexed pages and sent users back through links. AI systems can consume the information themselves, summarize it, and complete actions without generating a human page view.

Cloudflare has been building infrastructure around that new model. AIstify recently covered Cloudflare Wallets, which gives autonomous agents controlled stablecoin balances for purchasing APIs, data, and online services. The company has also developed systems for publishers to charge automated agents for content.

More Bots Also Mean a Larger Security Problem

Seifert warned that the rise in machine traffic will bring a corresponding increase in cybersecurity threats. A larger automated population means more legitimate agents, but also more scrapers, credential attackers, fraud systems, vulnerability scanners, and malicious bots capable of operating continuously at machine speed.

Modern agents also blur the old distinction between a browser used by a person and automated software. AI systems can operate real browsers, execute JavaScript, click interface elements, solve multi-step workflows, and imitate patterns that once provided reliable evidence of human activity.

Cloudflare has already responded with Precursor, a behavioral detection system designed to distinguish humans from increasingly capable bots by analyzing activity across an entire session instead of relying on a single CAPTCHA or request.

The company’s own data says automated bot traffic has already eclipsed human activity and reached roughly 57% of web requests. As agents become more capable, Cloudflare expects defenders to rely increasingly on behavior, cryptographic identity, permissions, and policy rather than simply asking whether a request came from a bot.

The 1,000-to-1 projection remains a forecast, not a measured outcome. Seifert acknowledged that Cloudflare’s earlier predictions underestimated how quickly machine traffic would grow, and the next five years could develop differently.

But the direction is increasingly visible in current data. People are not disappearing from the internet. Instead, each person may soon be surrounded by large numbers of automated systems browsing, negotiating, searching, buying, monitoring, and communicating on their behalf.

If Cloudflare’s forecast proves even approximately correct, the internet of the early 2030s will no longer be a network where machines primarily serve human page views. Human activity will remain important, but the majority of network interactions may happen between software systems that operate without a person watching each individual request.

Disclaimer: AIstify is an independent media brand owned and operated by NuvexMedia LLC, publishing news, research, and insights on artificial intelligence, emerging technologies, automation, and related industries. NuvexMedia LLC invests in and collaborates with companies across the AI, technology, software, and digital innovation sectors. These relationships do not influence AIstify’s editorial coverage, and the publication maintains full editorial independence to provide accurate, timely, and objective information. © 2026 NuvexMedia LLC. All rights reserved. This content is for informational purposes only and should not be considered legal, tax, investment, financial, or other professional advice.

AI & Machine Learning, News